Risk of hypoglycemia in users of human insulin. The Wisconsin Epidemiologic study of Diabetic Retinopathy.

Abstract

To evaluate whether there has been a change in frequency of hypoglycemic reactions associated with use of insulin of animal or human type. Data are from a longitudinal population-based study of people with diabetes onset before 30 years of age who were receiving primary care in an 11-county area of south central Wisconsin in 1980. Interviews of study participants were conducted regarding occurrence of hypoglycemic reactions and the frequency, dose, and type of insulin used. At the 4-year follow-up, 727 of the 765 subjects were using only animal insulin and 33 were using human insulin; by the 10-year follow-up, 352 were using animal insulin and 388 were using human insulin. Those using animal insulin reported fewer hypoglycemic reactions than did users of human insulin (chi (1)2 = 4.66, P = 0.03). Those who changed insulin type between visits were no more likely to report more hypoglycemic reactions during the year before the visit than those who remained on the same type at the two examinations (chi (4)2 = 3.37, P = 0.50). At the 10-year exam, users of human insulin were more likely to be taking multiple doses of insulin (86.6%) than were users of animal insulin (73.9%; P < 0.0001). In multiple logistic regression, the following variables were significantly related to the frequency of hypoglycemic reactions at the 10-year visit: body mass index (lower), glycosylated hemoglobin (lower), and being female. In this population, level of glycemia is related to the frequency of hypoglycemic reactions. While type of insulin (human vs. animal) may be related to glycemia, it appears to have little independent effect on the frequency of reactions.